org
Received: 04.04.2015
Published: 19.10.2015
ISSN: 21491402
Faculty of Letters and Humanities, Hay El Baraka Ben M'sik Casablanca B.P. 7951, University of Hassan II Casablanca, Morocco
2
Department of Mathematics, University of New Mexico,705 Gurley Avenue, Gallup, NM 87301, USA
Abstract  In this paper we present a new distance measure between neutrosophic refined sets on the basis of extended
Hausdorff distance of neutrosophic set and we study some of their basic properties. Finally, using the extended Hausdorff
distance and/or similarity measures, an application to medical diagnosis is presented.
Keywords  Neutrosophic sets, similarity measure, neutrosophic refined sets, extended Hausdorff distance.
1. Introduction
The neutrosophic set theory (NS) proposed in 1995 by Smarandache [6] was the generalization of the
(FS for short) [21], intuituionitic fuzzy set ( IFS for short) [20] and so on. In fuzzy set, the object,
partially belong to a set with a membership degree (T) between 0 and 1 whereas in the IFS represent
the uncertainty with respect to both membership (T [0, 1]) and non membership (F [0, 1]) such that
0 T + F 1. Here, the number I = 1 T F is called the hesitation degree or intuitionistic index. In
neutrosophic set, indeterminacy is quantified explicitly and truthmembership, indeterminacymembership and falsitymembership are independent. From scientific or engineering point of view, the
**
65
neutrosophic set and set theoretic view, operators need to be specified. Otherwise, it will be difficult
to apply in the real applications. Therefore, H. Wang et al [8] defined a single valued neutrosophic set
(SVNS) and then provided the set theoretic operations and various properties of single valued
neutrosophic sets. Many researches on neutrosophic set on neutrosophic set theory and its applications
in various fields are progressing rapidly [e.g., 2, 3, 4, 7, 8, 10, 11, 12, 15, 16, 17, 18, 19, 22, 23, 25,
31, 32, 34, 35, 36, 37, 38, 39].
The study of distance and similarity measure of NSs gives lots of measures, each representing specific
properties and behavior in reallife decision making and pattern recognition works. For measuring the
degree of similarity between neutrosophic sets, Broumi et al. [33] proposed several similarity measures
and investigated some of their basic properties. Ye [16] presented the Hamming, Euclidean distance
and similarity measures between interval neutrosophic sets. The generalized weighted distance and
similarity measures between neutrosophic sets were given by J.Ye [15]. Also, the same author
proposed three vector similarity measures for simplified neutrosophic sets (SNSs for short), including
the Jaccard, Dice, and cosine similarity measures for SVNS and applied them to multicriteria decisionmaking problems with simplified neutrosophic information. Therefore, S.Broumi et al. [42] extended
generalized weighted distance between neutrosophic sets (NSs) to the case of interval neutrosophic
sets. Hanafy et al. [13, 14] presented the correlation measure neutrosophic sets.
The multi set introduced by Yager [30] allows the repeated occurrences of any element and hence the
fuzzy multi set (FMS for short) can occur more than once with the possibly of the same or the different
membership values.
Recently, based on [8], the new concept neutrosophic refined set(neutrosophic multisets) NRS was
proposed by Broumi et al. [40] which allows the repeated occurrences of different truth membership,
indeterminacy and non membership functions. Later on, Broumi et al. [40] studied correlation measure
for neutrosophic refined sets and gave an application in decision making. The same author [41] defined
the similarity measure between neutrosophic refined sets based on cosine function. The concept of
NRS is a generalization of fuzzy multisets [42] and intuitionistic fuzzy multisets [43].
In this paper we extend the Hausdorff distance between neutrosophic sets to the case of neutrosophic
refined sets (NRSs).
The organization of this paper is as follows: In section 2, the neutrosophic sets, neutrosophic refined
sets and extended Hausdorff distance of neutrosophic sets are presented. The section 3 deals with the
proposed extended Hausdorff distance and similarity measure for neutrosophic refined sets. Section 4,
present an application of extended Hausdorff distance and similarity measure to medical diagnosis.
2. Preliminaries
This section gives a brief overview of concepts of neutrosophic set [6], and neutrosophic refined sets
[8], Hausdorff distance and extended Hausdorff distance between NSs [33].
66
A = {< x:
( ),
( )>,x X}
where the functions T, I, F : X ]0, 1+[ define respectively the degree of membership (or Truth), the
degree of indeterminacy, and the degree of nonmembership (or Falsehood) of the element x X to the
set A with the condition.
( )+
( )+
( ) 3+
(1)
From philosophical point of view, the neutrosophic set takes the value from real standard or nonstandard subsets of ]0, 1+[. So instead of ] 0, 1+[ we need to take the interval [0, 1] for technical
applications, because ]0, 1+[ will be difficult to apply in the real applications such as in scientific and
engineering problems. For two NS,
( ),
={ <x ,
And
={<x,
(1)
(2)
( )
( ),
( )> 
( )
if and only if
( )> 
( )
( )=
(2)
} the two relations are defined as follows:
( ), ( )
( ), ( )
( ) , ( ) = ( ), ( ) =
( )
( )
( ),
( ),...,
( ),
( ),,
( ))>: x
X}
( ),,
( ))
Where
( ), ( ),..., ( ): E [0 ,1],
( ), ( ),..., ( ): E [0 ,1], and
( ), ( ),..., ( ): E [0 ,1]
( )+ ( )+
such that 0
(
( ),
( ),,
( )
( ),
67
}
] and B=[
,
(
(3)
] in areal space R.
) between A and B satisfies the following properties
(D1) 1
(
) 1.
(
) =0 if and only if A = B; for all A, B
(D2)
(D3)
(
)=
(
).
(D4) If A B C, for A, B, C, then
(
)
(
and
*
( )
( ) 
( )
( ) 
( )
( )+
(4)
Where A = {< x, A(x), A(x), A(x)>} and B = {< x, B(x), B(x), B(x)>}.
The terms and symbols used in [5] are changed so that they are consistent with those in this
section.
Let X= {x1, x2, , xn} be a discrete finite set. Consider a neutrosophic set A in X where
TA(xi), IA(xi), FA(xi) [0,1] for every xi X, represent its membership, indeterminacy, and nonmembership values respectively denoted by A = {< x, TA(xi), IA(xi), FA(xi)>}.
Then, the distance between A NSs and B NSs is defined as follows.
(
*
( )
( )  ( )
( ) 
( )
( )+
(5)
(
)= H(A, B)denote the extended Hausdorff distance between two neutrosophic
Where
sets(NS) A and B.
Let A, B and C be three neutrosophic sets
(
) = H(A, B)=
*
( )
( )  ( )
( )
( )
( )+
(6)
68
*
H(A, C)=
( )  ( )
( ) 
( )
( )+
(7)
( ) 
( )
( )+
(8)
*
H(B, C)=
( )  ( )
Hung and yang [45] presented their similarity measures based on Hausdorff distance as
follows
Definition2.6 The similarity measure based on the Hausdroff distance is
(
(A,B)= 1(A,B)= (
(A,B)= (
Where
).
(
) (
))(
).
)).
Where
and B.
*
( )
( )  ( )
( ) 
( )
( )+
(9)
) = H(A, B)denote the extended Hausdorff distance between two neutrosophic sets A
We extend the above equation (9) distance to the case of neutrosophic refined set between
follows:
and
as
when the sets A and B do not have the same number of subcomponents for T, I, F. If
A = {<x,(
( ),
( ),...,
( ),
( ),,
( ))>: x
X}
B = {<x,(
( ),
( ),...,
( ),
( ),,
( ))>: x
X}
and
then we take m = max{a, b, c, p, r, q} and transform A and B into refined neutrosophic sets where all
components T, I, F have each of them m subcomponents, i.e.
69
A = {<x,(
( ),
( ),...,
( )), ( ( ), ( ),...,
( )), (
( ),
( ),,
( ))>: x
X}
B = {<x,(
( ),
( ),...,
( )), ( ( ), ( ),...,
( )), (
( ),
( ),,
( ))>: x
X}.
and
{
( )
( )  ( )
( ) 
( )
( )}} (10)
(D1)
(
) 0.
(
) =0 if and only if A = B; for all A, B NRSs.
(D2)
(D3)
(
)=
(
).
(D4) If A B C, for A, B, C NRSs, then
(
)
(
)
) and
Proof: (D1)
(
) 0.
As truthmembership, indeterminacymembership and falsitymembership functions of the NRSs lies
between 0 and 1, the distance measure based on these function also lies between 0 to 1.
(D2)
(i)
) =0 if and only if A = B.
( )
( ),
( ) = ( ) and
( ),  ( )
( ) and 
( )=
( )
) =0.
(
)=0.
( )
( ),  ( )
( ) and ( )
( )=0
( ),
( )
( ) and
( )
( )
(D3)
)=
70
It is obvious that
( )
( )
( )
( ), ( )
( )
( )
( )
and
( )
( )
( )
( )
But
( )
( ) =
( )
( ) , ( )
( )= ( )
( ),
and
( )
( )=
( )
( )
Hence
(
)=
=
=
{
( )
( )  ( )
( ) 
( )
( )}}
{
(
).
{
( )
( )  ( )
( ) 
( )
( )}}
Remark: Let A, B
( )
( ), ( )
( ),
(D4) If A B C, for A, B, C
(
)
Let A
( )
 if
NRSs, then
( )
( )
(
( )
( )
( )
(i)
( )
 ( )
 ( )
( )
 ( )
( )
( )
then
H(A,C)=
) and
( )for all xi X
 ( )
( )
( )
( )
( )
And 
(ii)
71
( )
( )
( )
( ) for all xi X
 ( )
( )
 ( )
( )
 ( )
( ) for all xi X
 ( )
( )
And  ( )
( ) 
 ( )
( )for all xi
( )
X
( ) for all xi
( )
and
( )
( )
( )
( )
( )
( )
( )
for all xi
{
( )
( )  ( )
{
{
{
( )
( )
{
( )  ( )
( ) 
( )}}
( )  ( )
( ) 
( )
{
( )
( ) 
( )
( )
( )}}
( )
( )}}
( )}}
( )  ( )
( ) 
That is
 If 
( )
( )
( )
(
( )
)AND
 ( )
(
( )
Then
H(A,C)= ( )
(a)
( )
( )
( )for all xi X
 ( )
And
( )
( )
( )
( )
( )for all xi X
72
 ( )
(b)
( )
( )
and
( )
( )
( )
( )
( )for all xi X
 ( )
( )
( )
( )for all xi X
( )
 ( )
 ( )
 ( )
( )
( )
 ( )
 ( )
( )and
( ) for all xi X.
{
( )
( )  ( )
{
{
{
( )
{
( ) 
( )
( )  ( )
{
( )
( )
( )}}
( )  ( )
( ) 
( ) 
( )
( )
( )}}
( )
( )}}
( )}}
( )  ( )
( ) 
That is
(
. If 
( )
( )
) AND
 ( )
(
( )
)

( )
(
( )
Then
H(A,C)=
(a) 
And
(b) 
( )
 ( )
( )
( )


( )
( )
( )
( )
( )
 ( )
 ( )
( )
( ) But
( )for all xi X
( )
( ) for all xi X
( )
( )for all xi X

And ( )
( )
( )
73
( )
 ( )
 ( )
( )for all xi
( )
( ) and 
( )
( )
( )
( )
( )
( ) for all
{
( )
( )  ( )
{
( )
{
( )
( )  ( )
{
( )
( ) 
( )
( )  ( )
( ) 
( )
( )  ( )
( )}}
( ) 
( )
( )}}
( ) 
( )
( )}}
).
( )}}
That is
(
From
) AND
( ( ) ( ))
(
(11)
74
It is well known that similarity measure can be generated from distance measure. Therefore we may use
the proposed distance measure to define similarity measures.
Based on the relationship of similarity measure and distance we can define some similarity measures
between NRSs A and B as follows.
Definition The similarity measure based on the extended Hausdorff distance is
(A, B)= 1(A, B)= (
(A, B)= (
(
(
)
)
) (
))(
)
(
))
(A, B)
S(A, C)
75
Table I 3valued neutrosophic set: The Relation between Patient and Symptoms
Q
Temperature
Cough
Throat
Head Ache
Body Pain
Let the samples be taken at three different timing in a day (morning, noon and night)
Table II neutrosophic refined set:The Relation between Patient and Symptoms
Q
Temperature
Viral Fever
(0.2, 0.5, 0.6)
Tuberculosis
(0.4, 0.6, 0.5)
Typhoid
(0.4, 0.3, 0.5)
Throat disease
(0.3, 0.7, 0.8)
Cough
Throat
Head Ache
Body Pain
Viral Fever
Tuberculosis Typhoid
Throat
disease
0.34667
0.366
0.426
0.2533
0.3
0.326
0.2733
0.346
0.34
0.22
0.266
0.3067
S=1d
Proposed similarity
Viral
measure(
(A,B)) Fever
0.6534
0.634
0.574
Proposed similarity
Viral
measure(
(A,B) Fever
0.4852
0.4641
0.4025
76
The highest similarity measure from table gives the proper medical diagnosis
Patient P1 suffers from Tuberculosis and Typhoid, Patient P2 suffers from Typhoid and
Patient P3 suffers Typhoid
5 Conclusions
In this paper we have presented a new distance measure between NRS on the basis of extended
Hausdorff distance of neutrosophic set, then we proved their properties. Finally we used this distance
measure in an application of medical diagnosis. Its hoped that our findings will help enhancing this
study on neutrosophic set for researchers. In future work, we will extended this distance to the case of
interval neutrosophic refined sets.
References
Q. Ansari, R. Biswas and S. Aggarwal, Neutrosophic classifier: An extension of Fuzzy
Classifer, Applied Soft Computing, 13,pp.563573, 2013.
[2] A. Kharal, A Neutrosophic Multicriteria Decision Making Method, New Mathematics and
Natural Computation, Creighton University, USA, 2013.
[3] C. Ashbacher, Introduction to Neutrosophic Logic, American Research Press, Rehoboth, 2002.
[4] D. Rabounski, F. Smarandache, L. Borissova, Neutrosophic Methods in General
Relativity,Hexis, 10, 2005.
[5] E. Szmidt and J. Kacprzyk, A Note on the Hausdorff Distance between Atanassovs
Intuitionistic Fuzzy Sets, NIFS ,vol. 15,no. 1, pp.112, 2009.
[6] F. Smarandache, A Unifying Field in Logics, Neutrosophy: Neutrosophic Probability, Set and
Logic, Rehoboth: American Research Press, 1998.
[7] F. G. Lupiez, On Neutrosophic Topology, Kybernetes, 37(6),pp.797800, 2008.
[8] Florentin Smarandache, nValued Refined Neutrosophic Logic and Its Applications to Physics,
Annual Fall Meeting of the American Physical Society OhioRegion Section, FridaySaturday,
October 45, 2013; Cincinnati, Ohio. http://meetings.aps.org/Meeting/OSF13/Event/205641
[9] H. Wang, F. Smarandache, Y. Q. Zhang and R. Sunderraman, Single Valued Neutrosophic
Sets, Multispace and Multistructure 4,pp.410413, 2010.
[10] Deli, S. Broumi and F. Smarandache, Neutrosophic Soft Matrices and NSMDecision Making,
Journal of Intelligent & Fuzzy Systems, DOI:10.3233/IFS141505,2015, in press
[11] I. Deli and S. Broumi, Neutrosophic Soft Relations and Some Properties, AFMI, vol.9, no.1,
pp.169182, 2015.
[12] I. Deli, Y. Toktas, and S. Broumi, Neutrosophic Parameterized Soft Relations and Their
Application, NSS, vol. 4, pp.2534, 2014.
[1]
77
78
[36] S. Broumi, I. Deli, and F. Smarandache, Neutrosophic parametrized Soft Set Theory and Its
Decision Making Problem, International Frontier Science Letters, 1 (1), pp.0111. 2014.
[37] S.Broumi and F Smarandache, On neutrosophic implications, NSS, vol. 2, pp. 917, 2014.
[38] S. Broumi, and F. Smarandache, Lower and Upper Soft Interval Valued Neutrosophic Rough
Approximations of An IVNSSRelation, Proceedings of SISOM ACOUSTICS, International
Conference, Bucharest, 2223 May 2014, pp. 204211.
[39] S. Broumi, M. Dhar, and F. Smarandache, Rough Neutrosophic Sets, Italian Journal of Pure
and Applied Mathematics, no. 32, pp.493502, 2014.
[40] S. Broumi and I. Deli, Correlation Measure for Neutrosophic Refined Sets and Its Application
In Medical Diagnosis, Palestine Journal of Mathematics, Vol. 5(1) (2016) , 135143
[41] S. Broumi, and F. Smarandache, Neutrosophic Refined Similarity Measure Based on Cosine
Function, NSS,vol.6, pp.4248, 2014.
[42] S. Broumi, I. Deli and F. Smarandache, Distance and Similarity Measures of Interval
Neutrosophic Soft Sets, Critical Review, Center for Mathematics of Uncertainty, Creighton
University,USA,8 pp.1431, 2014.
[43] S. Sebastian and T. V. Ramakrishnan, MultiFuzzy Sets, International Mathematical Forum,
5(50), pp. 24712476, 2010.
[44] T. K. Shinoj and S. J. John, Intuitionistic Fuzzy Multisets and Its Application in Medical
Diagnosis, WASET, vol.6, pp.0128, 2012.